The Translation and Cross-Cultural Adaptation of The Persian Version of The WERF EPHect Endometriosis Patient Questionnaire: A Cross-Sectional Study

Endometriosis is an inflammatory oestrogen-dependent chronic disease with a high prevalence among women of reproductive age, which is associated with negative impacts on fertility, quality of life, and functionality ( 1 , 2 ). This disease is as common as diabetes, as it is reported that 1 in 10 women are affected by it ( 3 ). About 30 to 50% of endometriosis patients also experience infertility impairment ( 4 ). Endometriosis is associated with a high comorbidity burden, increased use of medical resources, and excessive costs, especially for younger patients ( 5 ). Despite the particular importance of this disease, large prospective investigations in the field of endometriosis based on validated and standardized data are still lacking ( 3 ). Variations in the currently used methods of collecting and reporting data lead to large differences in the results of various studies, making it impossible to compare and/ or combine data to use for a meta-analysis ( 3 ). The World Endometriosis Research Foundation (WERF) is an international group of experts whose current goal is to standardize the collection of clinical and demographic information of endometriosis patients, as well as the collection and storage of biological samples, to facilitate joint and international endometriosis research. This foundation has developed several standardized questionnaires, which have been created with the collaboration of multidisciplinary experts ( 6 , 7 ). This action is designed as a project to coordinate accurate clinical and demographic data of endometriosis patients and the endometriosis bio-bank. It is hoped that this project will facilitate the conduct of multicenter clinical trials on a large international scale and with appropriate conditions, leading to the discovery of biomarkers and therapeutic targets to help advance re search in the field of endometriosis ( 8 ). In order to facilitate collaboration between endometriosis researchers, which is only possible through coordinated data collection, it is suggested that all researchers around the world use the standard clinical questionnaire of the endometriosis project ( 6 ). The Endometriosis Phenome and Biobanking Harmonisation Project (EPHect) clinical questionnaire is currently available in Arabic, Assamese, Chinese, Danish, Dutch/Finnish, French, German, Hindi, Latin American Spanish, Italian, Canadian, Malay, Mar athi, Portuguese, Spanish, and Turkish languages. Also, currently this questionnaire is being translated into Polish and Thai; however, it has not yet been translated into Persian. In addition, translation and cultural adaptation are necessary for the content and face validity of the questionnaire in each different language. So far, this research has been performed only in the Canadian, Turkish, and French languages, and there is no study available in the Persian language. The objective of the present study is to accurately translate, culturally validate, and assess the reliability of the Persian version of the standard endometriosis patient questionnaire (EPQ) for applications in epidemiological and clinical research on endometriosis in Iran.

In comparison of the Persian questionnaire and the original English questionnaire, no significant differences were observed, and only in some cases there were slight variations with regard to synonyms. In assessment of the pre-final version, the following items needed to be adapted to the Iranian culture and were changed: Question A4 (hormonal drugs): Medicines that we do not have access to in Iran or that do not exist in the Iranian pharmacopoeia drugs were removed from the Persian questionnaire. Question B4.1: Due to the fact that sperm donation is not allowed in Islamic religion, so the sperm donation item was removed from the relevant table. Question F2-F4: Ethnicity and ancestry are completely different in Iran, and we needed to change these things according to our culture and ethnicity. For this purpose, question F3 was completely removed and questions F2 and F4 were changed according to the Iranian population. Question F17: Alcoholic beverages are legally prohib ited in Iran, so the names of alcoholic beverages were removed from the Persian questionnaire. The answers provided by the participants to the questions inquired in the initial validation phase are as follows: i. The questionnaire was very long and boring, it took about an hour and a half (5 participants), ii. The questions were clear, except for the different type of hormonal drugs that I did not know about (2 participants), iii. Some parts of the questionnaire were confusing and it seemed to be repeated questions about the same topic. For example, it was confusing that every question was asked about three months ago and the whole life-time in same places in the table, which makes tables to be complex and the eyes tired. It would be better to ask questions related to the last 3 months and questions related to the subject’s lifetime separately (3 participants), iv. There were no questions about nutrition or environmental factors. It would be bet ter to ask about the type of nutrition, eating habits, the use of plastic or disposable food containers, the type of sanitary napkin used or exposure to radiation, etc. (2 participants), v. There were no questions about the amount of stress and the subject’s sleeping habits (1 participant), vi. The indiscriminate use of oral contraceptive pills or other hormones was not asked in the questions (1 participant), vii. It would be better to ask questions about anal intercourse during menstruation (1 participant), viii. It would be better to ask about having or not having meditation or prayer and religious worship as relaxation methods (2 participants), ix. It would be better to ask questions about cleaning methods after using the toilet (2 participants), x. It was better to ask questions about the place of residence and the level of environmental pollution and the climate of the habitat (1 participant), and xi. It would be better to ask questions about hookah smoking (2 participants). Regarding face validity, apart from the changes related to Persian grammar, the date of completion of the questionnaire was moved to the first page of the questionnaire, and the following items were modified and applied to the final version of the questionnaire due to the frequent mistakes of the patients: Question C17 (time of last sexual activity): A few days ago was added to the items. In the same question, the last item (avoidance of sex due to pain) was moved to the next line for a better under standing of the patient. Question C28 (last item of the pelvic pain severity scale): Visually, the shape of the lines has changed. Question C29.2: (last experience of pelvic pain), we added last week. Question C41 (pelvic pain intervention with the patient's social activities): separate lines were allocated for education and work and social or recreational activities due to the patients’ inaccuracy in the completion time. Question D4 (history of previous surgeries): cases such as the removal of the appendix, which is performed only once and cannot be repeated, were filled in black in the column of the number of surgeries. Some patients had confused ovarian removal surgery (oophorectomy) with follicle puncture during in vitro fer tilization (IVF) treatments, so its English equivalent, oo phorectomy, was added in front of this item. Question D10 (history of endometriosis in first-degree relatives): To increase the accuracy of patients’ responses, the proportions of each maternal and paternal family were specified in the column titles. Questions E1, and E2 (history of taking painkillers and other drugs): yes and no were defined separately. Question F10 (person’s body shape in each age catego ry): Due to carelessness in the answers, the photo and the corresponding numbers were moved to the bottom of the table. Demographic characteristics of the participants in the final validation stage (n=30) Thirty endometriosis patients with an average age of 33.8 ± 4 years and an average body mass index of 23.7 ± 3.6 kg/m2 participated in the study. In Table 2, the ethnic ity of the patients, their education levels, and their history of infertility and pregnancy records were reported. The survey shows that most of the patients were of Persian ethnicity (43.2%), and also most of them had an academic education (56.7%). Meanwhile, 63.3% of patients had no history of pregnancy, and 80% of the patients each had a history of primary infertility. Demographic and descriptive data of the participants (n=30) Values are presented as the mean ± standard deviation/error and number (%). BMI; Body mass index, and ART; Assisted reproductive technology. The results of the Kappa test showed that in most questions of the questionnaire, the percentage of agreement between the two tests had an acceptable level and was above the value of 0.4-0.9. Due to the large amount of data, the Tables S1-S5 (See Supplementary Online Information at www.ijfs.ir ) related to this part of the article are presented in the supplementary section, which is attached to the article. The first segment, which contains menstrual and hormonal history, includes 7 main questions. The response rate in questions A1 to A3 was 100%, but in A4 to A7, which was related to the use of hormones, there were cases of non-response to the questions, which remained as 398 missing in the data. The results of the Kappa test showed that in most questions of the questionnaire, the agreement between the two tests was acceptable and was above 0.6. The response rate of the participants, as well as the results of the Kappa test, are shown in Table S1 (See Supplemen tary Online Information at www.ijfs.ir ). The second segment is related to pregnancy or fertility and includes 4 main questions. Eleven patients (36.7%) had a history of pregnancy, and 19 patients (63.3%) had no previous pregnancy histories. The response rate of the other three questions of this section (B2-B4) was also 100 percent (Table S2, See Supplementary Online Information at www.ijfs.ir ). The third segment is appertained to pain and includes 43 main questions. The response rate of this section was lower than the other areas. However, the level of agree ment between the two tests regarding this part of the questionnaire was moderate to strong, and the range of Kappa values was between 0.6 and 0.9 (Table S3, See Supple mentary Online Information at www.ijfs.ir ). The fourth field is related to the patients' medical his tories and includes 10 main questions. In this section, the participation rate was high, and 100% of the questions were answered (Table S4, See Supplementary Online Information at www.ijfs.ir ). The fifth area is related to the patients’ drug histories and includes two main questions. The response rate regarding the main questions was 100%; however, in answering the sub-questions related to each medication, the response rate was decreased (Table S5, See Supplementary Online Information at www.ijfs. ir ). The sixth area is related to personal information and lifestyle of the patients and includes 19 main questions that had a high percentage of participation in answering (Table S6, See Supplementary Online Information at www.ijfs.ir ).

The EPHect clinical questionnaire is currently available in Arabic, Assamese, Chinese, Danish, Dutch/Finnish, French, German, Hindi, Latin American Spanish, Ital ian, Canadian, Malay, Marathi, Portuguese, Spanish, and Turkish languages. Also, this questionnaire is being translated into Polish and Thai languages, but it has not been translated into Persian yet. The present study is the first study that has accurately translated this questionnaire into the Persian language and adapted it to the Iranian culture. In general, our study showed that the Persian version of the WERF-EPHect questionnaire is practical and accept able to endometriosis patients. After the completion of the translation process and the preparation of the pre-final version, changes including the removal of some hormonal drugs and a number of pain killers (due to their unavailability in the Iranian pharma copoeia), the removal of items such as donated sperm and the names of Alcoholic beverages (due to incompatibility with Iran's Islamic culture), and modifying the classification of race and ethnicity according to Iran's ethnic struc ture were necessary. The results of the pilot study based on this pre-final version showed that about 70% of the participants com plained about the length of the questionnaire, since it had taken them about 1 to 1.5 hours to complete the questionnaire. About 71% of the study participants found the questionnaire instructions clear and easy to understand; however, they thought that it was difficult for them to complete the questionnaire at home, as it was not possible to ask questions from the researchers. Therefore, some items of the questionnaire remained incomplete due to misunderstanding or insufficient knowledge of the patients, which were mainly seen in the section of hormonal medication use history. In addition, more than 50% of the women complained about the tables in the questionnaire for being confusing. Due to the lack of familiarity with this type of table, which generally asks questions related to different stages of lifetime, it is probably helpful to use the assistance of the treatment staff to complete these items, or it is necessary to present the items in the form of an online questionnaire. Although the questionnaire demands a large amount of detail, the success in identify ing endometriosis requires obtaining this important data regarding pain, symptoms, menstrual status, reproductive and medical histories, as well as the demographic and lifestyle information. To promote successful completion of the questionnaire, we made the skip patterns more visible by using format ting techniques, such as highlighting the relevant font, in the text. However, some participants, even with a high level of education, were careless about the skip points and therefore we recommend that while completing the form, the participants should be informed about the skip patterns to prevent unnecessary questions from being completed by the researchers. Another method is to use brackets and parentheses or to change colors (if there is no color print limit) to separate the skip pattern. Further more, the usage of an electronic online platform to man age the questionnaire may automatically perform the skip function depending on the client’s reply, and on the other hand, the questionnaire data may be used for analysis with a greater accuracy and at a higher speed. This study con firmed the cultural validity of the Persian version of the WERF-EPHect. Questionnaire validity is very important because it ensures that the obtained results are an accurate representation of what we intend to measure. We were concerned that our study sample, which was collected at an infertility treatment center, may not be representative of the average Persian-speaking Iranian women with endometriosis, but our population was diverse in terms of age, clinical presentation, education, as well as marital and financial status. therefore, it was in fact a representa tive population suitable for different strata of Iranian so ciety. The significant strength of this study was the inclusion of women with endometriosis at different ages. This increased the diversity of the history of the participants with endometriosis, as well as diversity in their approach to the semantics and the concept of the questionnaire. The education level of the participants was an important factor in the participants’ understanding of the medical terms in the questionnaire. In addition, we noticed that the participants with a lower level of education seemed to pay less attention to the instructions at the beginning of the questionnaire, so the presence of the researcher next to the people completing the questionnaire seemed necessary to provide possible explanations. In order to avoid data errors, participants’ contact information should be collected in a separate section of the questionnaire. In comparison to the previously published articles, explaining their translation and cross-cultural adaptation of the WERF EPHect EPQ profoundly, we realized that our method and findings are equivalent. The present study in cluded the highest recommended number of participants. The results of the present study are in accord with previous studies in other parts of the world. In the study by Dimentberg et al. ( 3 ), for example, a culturally valid French Canadian version of the EPHect Endometriosis Patient Questionnaire-Standard (EPQ-S) from the WERF was developed. The results showed that a total of 22 women agreed to participate in the study, and ultimately, 17 of them participated in the final validation. These women were in pre-menopausal age and had symptoms of pelvic pain or suspected endometriosis, and had been referred to the women's health clinic in Quebec City, Canada. In the first visit of these subjects to the clinic, they completed the French version of the questionnaire, and one week later, its cultural validation was performed. The average age of the participants was 38.9 years, and 15 (88%) had a confirmed diagnosis of endometriosis. Fourteen participants (82%) found the questionnaire clear and relevant. The average time to complete the questionnaire was 36.25 ± 10.8 minutes. Five participants (29%) considered the questionnaire to be long. The validation process resulted in the addition of lines at the end of each section to provide space for specific comments and some clarifications regarding the timelines ( 3 ). Similarly, the Danish trans lation of the EPHect EPQ-S questionnaire and its cross cultural adaptation, as well as the equivalency of the electronic version of the questionnaire were evaluated by Thomsen Holdgaard and his colleagues. For this purpose, 10 patients with confirmed diagnosis of endometriosis through ultrasound, laparoscopy or histology were in cluded in the study to evaluate the written questionnaire, and 5 patients for the electronic format through the Dan ish Endometriosis Association or the endometriosis clinic at Copenhagen University Hospital. The reported median time-to-complete was 62 minutes (range: 29-110). In general, it was easy for 50% of the patients to understand the questionnaire. About 80% of the patients described the questionnaire as long and time-consuming, but only one participant considered it too long (10%). 60% of the patients stated that they had problems with remembering when answering the questions that were related to the history of their adolescence years or that they could not an swer the questions, in which many details were required. Finally, the researchers concluded that the Danish version of the EPQ questionnaire, both in paper and electronic forms, was similar and comparable to the original English questionnaire. However, it is argued that questions of measurement units, ethnicity, and education systems should be addressed before cross-country comparisons can be made ( 11 ). In another study, Mis et al. ( 12 ) in vestigated the Turkish translation and cultural adaptation of a shorter version of the EPHect EPQ questionnaire. The reported duration of questionnaire completion ranged from 30 to 60 minutes. Also, it was reported that most of the participants found the questionnaire to be clear, but all stated that the questionnaire was too long. 40% of the participants commented on the length of the questionnaire after completing part D, and the rest of the patients commented that the questionnaire was long after completing the entire questionnaire. Despite the fact that jumping patterns were included in the questionnaire, 10% of the par ticipants ignored it. Consequently, we enhanced the visual distinction of the skip patterns within the questionnaire and recommend that the participants be informed about these patterns prior to beginning the questionnaire. Notably, 20% of the participants with primary or secondary education reported that certain questions appeared repetitive, specifically, questions regarding pelvic pain (section C2), pain during or after vaginal intercourse (section C15), and general pelvic pain (section C27). When research as sistants clarified the differences among these sections, it became evident that the participants had not adequately read or comprehended the initial instructions for these sections. Therefore, the researchers improved the visual clarity of the instructions and advised participants to thoroughly review the instructions for these sub-sections before beginning the questionnaire ( 12 ). Finally, as the results of the previous studies were mentioned earlier, except for one study that examined the shortened version of the questionnaire, in the rest of the studies similar to the present study, the average time required to complete the questionnaire was reported to be approximately one hour. Also, in all of the recent studies, most of the patients mentioned that the questionnaire was too long and some questions were repetitive. In agreement with previous studies, in our study, most of the patients mentioned the length of the questionnaire, and some patients did not notice the skipping pattern in the study, and they did not answer all the questions in section C of the questionnaire. Most of the patients stated that too many details were asked in the questionnaire, some of which were questions related to their adolescence age that they could not remember, so most of the patients left the questions related to the severity of pelvic pain in their adolescence age unanswered. The time interval between the test and the retest is one of the challenges when using the "test-retest" reliability method of the scales of a questionnaire. In the pre sent study, the time interval of two weeks is considered to avoid forgetting of the terms of the questionnaire for the participants. On the other hand, based on our observations, no change occurred in the measured variables, and the high values of the Kappa coefficient showed that the Persian questionnaire has a high level of reliability. Currently, the EPQ-S questionnaire is used by various centers around the world to examine pain, menstrual and fertility history, history of hormone use and infertility, and demographic characteristics and lifestyle in women diagnosed with or suspected of endometriosis. Validation of the EPHect EPQ-S in different languages and cultures allows for standardized and high-quality assessment of women with endometriosis, comparison between studies, and aggregation of the results obtained nationally and internationally. Collaborations of this nature result in studies that are demographically diverse and exhibit high levels of reliability and validity, thereby advancing significant research and discoveries for the millions of pa tients affected by endometriosis globally. Clinical trials that utilize non-standardized questionnaires for data collection frequently encounter issues related to the validity and comparability of their findings ( 3 ). The evaluation of pain symptoms is associated with challenges such as subjectivity, which is used in the EPHect EPQ-S question naire, a valid tool of visual analog scales. In the present study, a paper version of the questionnaire was used; it is suggested that an electronic version be designed to make it easier for patients to answer and access it.

The present study is the first study that has translated this questionnaire into the Persian language and adapted it to the Iranian culture. In general, our study showed that the Persian version of the WERF-EPHect questionnaire was practical and acceptable to patients, and it is a valid and reliable tool. Small changes related to jump points, explanations related to drugs and hormones are necessary for the patients’ better understanding. Also, the presence of a research assistant at the time of completing the questionnaire by patients will help to solve possible issues. Validation studies in different languages and cultures prevent the differences in those languages and cultures from creating obstacles for future collaborations and in ternational research on endometriosis. In addition, the opportunity to compare and aggregate the results of various studies at national and international levels is provided.

In this cross-sectional study, a total of 37 women aged between 18 and 45 years, who underwent diagnostic so nography or laparoscopy for endometriosis, or had symptoms of dysmenorrhea, dyspareunia, pelvic pain unrelated to menstruation for more than 6 months, were assessed. The evaluations took place from May 2021 to December 2023. The study protocol was approved by the Institutional Review Board and the Ethics Committee at Roy an Institute (IR.ACECR.ROYAN.REC.1400.156). EPQ clinical questionnaire contains 84 questions, which are divided into 6 sections with 14 questions on average in each section ( 6 ). The study participants were selected as an available sampling from the women’s infertility clinic and the ultra sound department of Royan Institute. Women who agreed to complete the questionnaire on the first visit prior to so nography at the clinic were included, and the validation took place approximately two weeks later. Women with a diagnosis of malignancy or any cognitive impairment hindering questionnaire completion were not included in the study. The research process was structured into two distinct phases: translation and cross-cultural adaptation, fol lowed by cross-cultural validation. The translation and adaptation of the Persian version of WERF EPHect were conducted in accordance with the established guidelines ( 9 ). The entire process was executed in the following sequence: i. Ensuring conceptual equivalence, ii. Conduct ing forward translation, iii. Performing backward transla tion, iv. Preparing the Persian pre-final version by an expert committee, v. Conducting cognitive interviews with the target group, and vi. Finalizing with proofreading. Conceptual equivalence: the phase consisted of per forming a literature review on endometriosis, as well as health and well-being concepts in the target language database. Two gynecologists were involved in ensuring the clarity and relevance of all concepts and questions in the WERF EPHect, as well as making any necessary adjustments to avoid cultural insensitivity For this purpose, after obtaining permission and receiving the instructions of the questionnaire from its developers on the translation of the EPQ questionnaire, the translation process was carried out according to the standard guideline ( 3 ). In the first stage, two researchers who had sufficient knowledge on the subject of the study (endometriosis) and were able to perform translation, translated the questionnaire into Persian independently. The translators were fluent in English, while their native language was Persian. The two translators combined their translated versions of the questionnaire with each other after resolving the differences. As an output of this stage, in addition to the combined version of the translated questionnaire, a written report on how to resolve the disagreements was also prepared (consensus opinion of the two translators). The final combined translated version (from the previ ous stage) was next translated from Persian to English, which was the original language of the questionnaire. The translation was done independently by two different experts, who were fluent in both English and Persian. In this stage, unlike the previous stage, the two translators were not experts in the relevant scientific topic (endometriosis). These translators had not participated in the previous stages and had no access to the original English questionnaire. The translation was done conceptually and not only word for word. The two translators along with the observer of the process combined the two translated versions of the questionnaire with each other after resolving the differences. As an output of this stage, in addition to the combined version of the translated questionnaire, disagreements were resolved through the consensus of the two translators. At this stage, meetings were held under the supervision of the responsible researcher and coworkers to identify and eliminate inappropriate expressions or concepts in the translated version and to check any differences be tween the original and the translated questionnaire. The four previously mentioned translators, researchers in volved in the study, methodologists, and one language expert studied all the documents and reports of the previous stages, as well as the written reports regarding the resolution of disagreements. Additionally, the group communicated with the designers of the original questionnaire and exchanged opinions. Then, considering all the above, the pre-final version of the Persian questionnaire was prepared. In this stage, to determine the acceptability and compre hensibility of the translated questionnaire and whether the patients will be able to complete the questionnaire alone and without requiring help from the health staff, a pre test was conducted on 7 patients with endometriosis, who were informed about the study prior to the test and verbal consent was obtained from them. Due to the large number of questions, the questionnaire was given to the patients to fill it out at home. After completing the questionnaire, the patients were called by one of the co-authors, and the items that they could not understand, as well as their opinions on the questions, were asked. The women who completed the questionnaire were con tacted at most three times, and if they did not respond or did not attend, they were excluded from the study. The following questions were asked of the pre-test participants: How long did it take to fill out the questionnaire? Is it too long? Are all questions clear? Are all questions relevant? Do you have any suggestions for modification? Do you have any other comments about the question naire? At this stage, the answers and suggestions that the patients provided in the initial pilot test were informed to the developer (World Endometriosis Society) on Febru ary 10, 2022, and the association's response was received. Finally, after confirming and applying the collected opinions, the final version of the Persian questionnaire was prepared and entered the validity and reliability phase. Considering that the validity and reliability of the tool may be affected during the translation process, there fore, after the translation, the following steps were taken to confirm the validity and reliability of the tool. Due to the standardization of the translated tool, only qualita tive validity checking is efficient. In order to examine the content validity of the questionnaire, the translated version of this questionnaire was evaluated by 6 experts (infertility fellowships and academic staff members familiar with endometriosis disease with more than 10 years of work experience). In this part of the study, these experts were asked to present their corrective views in writing regarding the grammar, using appropriate words according to the Iranian culture, placing the items in their proper places and giving appropriate points, after studying the tool carefully. The items were edited according to the suggestions and recommendations of the experts. The reliability of the final version of the Persian EPQ questionnaire was evaluated by the "test-retest" method. For this purpose, the final adjusted questionnaire was made available to 30 women on two occasions with an interval of two weeks, and reliability in repeatability was assessed separately for each question using the Kappa statistics. For questions with binary classification, Cohen's kappa was used, and for rating scales with three categories or more, we used weighted kappa. The kappa coefficient is a variable between 0 and 1 and is expressed as a percentage, and the minimum acceptable value of the Kappa coefficient is higher than 0.4. Values higher than 0.6 are good, and values higher than 0.8 are ideal ( Table 1 ) ( 10 ). Interpretation of Cohen’s Kappa